402 



Fishery Bulletin 90(2), 1992 



separation of sexes would have reduced variance in the 

 length-frequency distribution of the older age-classes 

 due to growth differences between sexes, there was 

 still considerable size overlap of age-classes and diffi- 

 culty in identifying modes after age 2 (Fig. 10). There 

 was a clear progression in the length of age-classes 1 

 and 2 through the year, although age-class-2 fish of 

 both sexes were somewhat larger than expected in 

 February based on the progression in length of this age- 

 class in subsequent months. 



Mortality 



The relative abundance of each age-class by sex was 

 determined for the five periods sampled in 1983 (Fig. 

 11). A line was fitted by least squares to the descend- 

 ing limb of the catch curve. Fish not considered to be 

 fully recruited to the sampling gear (circled points) 

 were excluded. There was no significant difference in 

 the slopes of the lines for males and females (ANCOVA, 

 F 0.85, df 1, 47, P 0.36) and no significant sex effect 

 (ANCOVA, F 1.23, df 1, 47, P 0.27) so a catch curve 

 was fitted to the combined data. The instantaneous rate 

 of annual mortality (Z) for males and females was 

 estimated to be 0.98 (SE 0.076). 



Discussion 



Lutjanus vittus was caught at depths of 20 m (the 

 shallowest depth sampled) to 120 m, with larger fish 

 tending to inhabit deeper waters. This tendency has 

 also been observed in other shallow-water lutjanids 

 such as L. aya (Moseley 1966), L. griseus (Starck 1971), 

 and L. hohar (Wright et al. 1986). 



Most ageing studies on lutjanids have relied on oto- 

 liths as the principal structure (see review by Manooch 

 1987).However, a few authors (i.e., Reshetnikov and 

 Claro 1976, Pozo and Espinosa 1982, Claro 1983, 

 Palaz6n and Gonzalez 1986) have used urohyals. 

 Reshetnikov and Claro (1976) had difficulty determin- 

 ing the boundaries of the annual increment after the 

 second or third annulus in urohyals because the annuli 

 were made up of multiple bands. It was our experience 

 that, despite this problem, increments were still easier 

 to measure on urohyals than on whole otoliths, and 

 preparation was far less time-consuming. 



Our preliminary investigation indicated that the same 

 number of checks were formed on a variety of hard 

 structures, including urohyals. Data on marginal in- 

 crements in urohyals showed a seasonal pattern with 

 one check being formed each year, consistent with most 

 other studies on lutjanids in tropical waters. However, 

 studies on two lutjanid species from Cuban shelf waters 

 have suggested that checks are formed twice a year 



01 23456789 



0123456789 

 Age (years) 



Figure 1 1 



Logj frequency against absolute age for male and female 

 Lutjanus vittus. Data for fish <2 years old (circled) were ex- 

 cluded from the regressions, as they were incompletely 

 recruited age-classes. 



(Espinosa and Pozo 1982, Pozo and Espinosa 1982). 

 The growth of male and female L. vittus was sig- 

 nificantly different after 3 years of age, with females 

 growing markedly slower than males. There are few 

 documented cases of growth rates differing between 

 sexes in lutjanids. However, female L. vittus in New 

 Caledonia were found to grow at a slower rate, and 

 slight growth differences were found in L. amabilis 

 in New Caledonia (Loubens 1980) and L. synagris 

 in Trinidad (Manickchand-Dass 1987). All mature 

 females observed were 3 years of age or older (unpubl. 

 data), and it seems likely that females grow more 

 slowly than males at this stage because they expend 

 proportionally more energy on gamete production than 

 do males. Stunting in females from a sexually preco- 

 cious population oiLates calcarifer was also attributed 



